MercadoLibre CEO: Agentic AI Will Kill Apps

The CEO's prediction is rooted in a broader industry shift away from app-centric interactions towards "conversational commerce." Instead of navigating menus, users will state their intent, and an AI agent will orchestrate tasks across multiple services in the background. This creates a new paradigm of "headless apps" that function purely as services for AI agents, not for direct human interaction. This vision extends to new hardware, with a wave of AI-native devices designed to replace the smartphone. Gadgets like the Humane AI Pin and Rabbit R1, along with forthcoming AI-integrated glasses, earbuds, and rings from major tech players, are built for a screenless, voice-first world. These devices rely on specialized chips and new operating systems built from the ground up for agentic AI control and contextual awareness. On the development side, AI is already deeply embedded in coding workflows, with some reports indicating that nearly 90% of software teams use AI. AI coding assistants are used for generating new code, creating tests, and assisting in code reviews. This has led to measurable increases in development speed, though it also raises concerns about potential decreases in code quality if not properly managed. While MercadoLibre is leveraging AI for developer efficiency, the company also plans to create 28,000 new jobs, with 3,300 in technology, suggesting a strategy of augmentation rather than wholesale replacement. The company's focus is on scaling operations and enhancing logistics, with AI as a core component of future growth. This aligns with a broader industry view that AI will augment developers by handling repetitive tasks, allowing them to focus on more complex problem-solving. The move towards agentic AI is already visible in e-commerce through tools that guide shoppers with natural language, turning search into a sales-like conversation. For its part, MercadoLibre has been using AI for personalized recommendations, fraud detection, and logistics optimization. CEO Marcos Galperin envisions an AI assistant within the company's fintech arm, Mercado Pago, that will help users manage bills and automatically invest surplus income. The transition to an agent-driven ecosystem requires a fundamental shift for builders. The focus moves from developing standalone applications to creating intelligent agents and APIs that can seamlessly connect to a central AI system. For developers, this means adapting to AI-powered IDEs like Cursor, which alter the workflow from line-by-line coding to high-level planning and instruction. Gartner forecasts that by the end of 2026, 40% of enterprise applications will embed AI agents, a significant increase from just 5% in 2025. This rapid adoption signals a critical window for companies to define their agentic AI strategy or risk falling behind. The shift is not just about new features, but a reimagining of the user experience to be more collaborative and proactive.